Noninvasive Quantitative Compression Ultrasound Central Venous Pressure: A Clinical Pilot Study

Objective: This is an initial study to validate central venous pressure (CVP) measurements derived from quantitative compression ultrasound (QCU). Impact Statement: This study is the first gold standard invasive validation of CVP estimation from QCU. Introduction: QCU finds the collapse force-the force required for complete occlusion-of the short axis of the internal jugular vein (IJV) to estimate CVP. Methods: We captured QCU data as well as the noninvasive clinical standard jugular venous pulsation height (JVP) on cardiac intensive care unit (CICU) patients at Massachusetts General Hospital (MGH). We compared these data to ground truth invasive CVP data from the MGH CICU. Results: Using linear regression, we correlated invasive CVP with collapse force (r ²: 0.82, error: 1.08 mmHg) and with JVP (r ²: 0.45, error: 1.39 mmHg). To directly compare our method to JVP, we measured the percentage of patients whose uncertainty estimates for QCU methods and for JVP overlapped with their invasive CVP counterparts. We found that the CVP overlap accuracy of collapse force (77.8%) and of collapse force and hydrostatic offset (88.9%) are higher than that of JVP (12.5%). Finally, we input QCU image segmentation data of the short-axis cross-sections of the IJV and carotid artery into an inverse finite element model to predict the invasive CVP waveform. Conclusion: These results validate the noninvasive technique for estimating CVP, namely, QCU, indicating that it may provide a desirable, middle-ground alternative to invasive catheterization and to visual inspection of the JVP.